OpenAI to add invisible watermark to ChatGPT and Codex text in the EU
The textGrain method shapes word choices so a detector can identify AI-generated text; API developers elsewhere can opt in.

OpenAI said on Monday it will add an invisible watermark to text generated by ChatGPT and Codex in the European Union to comply with the EU AI Act's transparency rules, which took effect Aug. 2, TechCrunch reported.
The watermark will roll out over the coming weeks to eligible users on all plans in the EU. Developers using OpenAI's API anywhere can turn it on for select models; it is off by default, and OpenAI said it is not making text watermarking a global default.
The method, called textGrain, subtly shapes the model's word choices to leave a pattern a detector can pick up. In OpenAI's tests, replacing 10 per cent of words with synonyms dropped detection from about 92 per cent to 66 per cent.
OpenAI said it will initially give detector access only to approved researchers and expert organizations, and cautioned that a missing watermark does not prove human authorship.
From ARQS Sponsored
Readers looking to act on this kind of technology shift can start with SignDesk, Northline course catalogue, ARQS security and Live Intelligence. In the ARQS Store, ARQS InvoiceFlow Billing Control Studio, ARQS InvoiceFlow Collections Console, AICL 104 — Regulatory Informatics & Unstructured Data are among the newest tools available.
